Recognition of isolated musical patterns using Hidden Markov models

Abstract

This paper presents an efficient method for recognizing isolated musical patterns in a monophonic environment, using Discrete Observation Hidden Markov Models. Each musical pattern is converted into a sequence of music intervals by means of a fundamental frequency tracking algorithm followed by a quantizer. The resulting sequence of music intervals is presented to the input of a set of Discrete Observation Hidden Markov models, each of which has been trained to recognize a specific type of musical patterns. Our methodology has been tested in the context of Greek Traditional Music, which exhibits certain characteristics that make the classification task harder, when compared with Western musical tradition. A recognition rate higher than 95% was achieved. To our knowledge, it is the first time that the problem of isolated musical pattern recognition has been treated using Hidden Markov Models.